Loss of torque with EFI?

[pastry_chef]

Interesting, if there could be a redo test with 2 different sizes of injectors per cylinder. Then shutdown the larger injector during low RPM.

[andyf]

In a recent article on the Hot Rod website a big BB Chevy is tuned with both a Dominator and a Holley Sniper system. It took them several days to dial in the fuel curve on the Dominator but the Holley Sniper learned its curve in just a few dyno pulls. The peak numbers were basically the same but the Sniper was down about 50 ft-lbs at the bottom of the RPM range. The article didn't mention this torque gap but I've seen the same thing happen when testing EFI vs. carbs. My guess is that it has to do with fuel mixing but I don't know if that is true or not. We don't know all of the other variables so it is just a guess what happened with that engine. For example, I assume the timing curve was the same but it might not have been. Same with the weather conditions, type of fuel, etc.

Has anyone else seen a drop in torque with EFI even though the peak numbers stayed the same? The Sniper is a throttle body type of setup so in theory the fuel mixing should be the same but maybe the injectors don't do quite as good a job as the annular boosters in the Dominator?

https://www.hotrod.com/articles/fuel-sc ... ow-to-fix/

I have seen this many times with EFI throttle body systems which learn to a specific air/fuel ratio everywhere.
Generally, an engine will benefit from a richer condition near peak torque and a slighter leaner condition at and after peak horsepower.
A tuner who can vary the air/fuel ratios at near those places will not see very much difference at all.

That is a good point. I typically lean out my EFI tunes a tad after peak hp and have them richer at peak torque. It is simple enough to do with EFI since you just adjust the table to make it happen. The race engine in my article was tuned that way which is probably one reason why it slightly better peak numbers than the carb. It still had a large loss at the bottom end though.

I don't think EFI has to have a torque loss since the OEM's don't have this problem. My guess is that there is just something in these combinations that created it.

[andyf]

Interesting, if there could be a redo test with 2 different sizes of injectors per cylinder. Then shutdown the larger injector during low RPM.

Yes that might be interesting. The OEM's used that idea for a while but they seemed to have gotten past it so I'm guessing they came up with another solution. They might have developed a more sophisticated control system for the injectors which the aftermarket tuners don't have yet.

[peejay]

Most of the guys I know making serious power on EFI would not consider a self learning system because the main benefit to the extra cost of EFI is being able to tune to a specific application. A self learning system can only go as far as what has been programed by someone else.

These EZ self learn setups arent real multi point EFi systems, basically you bought yourself an electronic carburetor...

With worse atomization.

I wonder if the difference at low speed is that, with a carb, it's flowing a low flow of more or less constant emulsified fuel, while the TBI system is flowing a small duty cycle of high flow droplets. Fuel quality suffers, even though quantity is "correct", is what I'm getting at.

That really should be the only difference between the two, anyway. The fuel is entering the airflow at the same point.

[peejay]

In a recent article on the Hot Rod website a big BB Chevy is tuned with both a Dominator and a Holley Sniper system. It took them several days to dial in the fuel curve on the Dominator but the Holley Sniper learned its curve in just a few dyno pulls. The peak numbers were basically the same but the Sniper was down about 50 ft-lbs at the bottom of the RPM range. The article didn't mention this torque gap but I've seen the same thing happen when testing EFI vs. carbs. My guess is that it has to do with fuel mixing but I don't know if that is true or not. We don't know all of the other variables so it is just a guess what happened with that engine. For example, I assume the timing curve was the same but it might not have been. Same with the weather conditions, type of fuel, etc.

Has anyone else seen a drop in torque with EFI even though the peak numbers stayed the same? The Sniper is a throttle body type of setup so in theory the fuel mixing should be the same but maybe the injectors don't do quite as good a job as the annular boosters in the Dominator?

https://www.hotrod.com/articles/fuel-sc ... ow-to-fix/

I have seen this many times with EFI throttle body systems which learn to a specific air/fuel ratio everywhere.
Generally, an engine will benefit from a richer condition near peak torque and a slighter leaner condition at and after peak horsepower.
A tuner who can vary the air/fuel ratios at near those places will not see very much difference at all.

Interesting.... is this also the case when you have electronic timing control? Because it also sounds like a way to alter the flame front speed.

[Tuner]

Don't the OE TBI systems have a water or exhaust heated heat exchanger between the TB and the manifold to heat/vaporize the fuel?

I recall a personal experience with a carb spacer on a '66 289 Ford where removal of the sharp edge that was upward-facing into the flow caused a huge loss of low end torque (added over a second to the 1/8th mi. time) and tanked the mileage from 25 to 15. No joke, no BS.

[modok]

I'm going to say it again.
The EFI guys think that, with the MAGIC of a fuel injector, all you need is a big throttle plate, and all the features that carburetors have is crutches to aid booster signal.
NO
Vac secondares=variable inlet manifold
Venturi=atomizarion booster

[Mike Laws]

If that were the case then Indy cars, Formula 1 cars, Nascar cup cars and almost all other venues of racing in the world use EFI. Why, I guess as many Carb nuts surmise they just haven't heard about how good carbs are yet ! HA

I'm pretty sure the classes mentioned require EFI. (not sure about F1.) It took a few modifications to the throttle body for the Nascar cup engines to make the same power as carburetors, which were size restricted. In the drag racing classes that allow either - carburetors are the overwhelming choice.

Having said all of that, there are venues that EFI is easier to make work. High lateral g classes and boosted come to mind...

[scott woodward]

Maybe there’s a lot of difference in air speed at lower rpm between the carbi and sniper.
The Dom sniper is a big unit.

[hoffman900]

I think part of the problem is people put EFI on applications that are optimized for a carburetor. An intake tract optimized for EFI will look nothing like a common wet plenum design, and the system is designed around that.

A Holley Dominator based carburetor is about as optimized as a carburetor as you’re going to get. So I don’t know why these magazines compare them to what is obviously a self learning TBI electronic carburetor.

As for other series, NASCAR and NHRA PS have mandated the type and where the throttle body has to be, the type of injectors, their locations, etc. Not really an ideal situation.

Where there are no rules to dictate it, and where the collective talent pool knows how to deal with it (not typical in the drag racing builder ranks), EFI prevails.

Another thing to consider, is outside dyno specific challenges, most engines don’t perform on a static dyno, and things like dealing with g loads, transient response, shift recovery all matter, and that’s where things start separating themselves.

[Mike Laws]

engine master did a test efi vs carb

1000CFM efi TB on a single plane with port bungs and port FI.

750CFM carb and 1050CFM carb

the 750 was a slighty better overall, but peak EFI was ahead due to the more restrcition in the 750 carb but at some rpm they were identical but the curves generally followed within 5 numbers (450hp engine). the efi had better AFR but the temp just below the TB was 40F lower (80f vs 40f) across the board lower due to vapourisation with carb.

with 1050 carb the carb was ahead overall by a consisten 6-8 numbers.

ignition timing was the same for all

so the results are expected when airflow is the same (same TB "flow") and AFR is the same the carb has better cooling of the charge and so improves slighty overall by a couple %.

if you are seeing 50ftlb difference something is wrong

rarely do you see efi with efi manifold vs a carb with single plane or similar which is the proper test. from a blank sheet of paper a performance efi manifold is not going to look like a single plane. if people were to test a carb on an efi manifold people would get their panties in a bunch.

The engine masters show is nothing like the engine masters competition.

When a competitor goes with a carb over efi, they have done everything they can to get the best score..30 or 40 pulls using &comparing both for averages and peaks.

Carbs are still the clear winner for wot

I've got a little over 3 years of experience with this unit: https://www.knfilters.com/eci They've been blasted by EFI advocates about it online, but it works very well. It makes more power and has been quicker on the track in every application I've run it on and I hear that some sharp tuners are using it but keeping results to themselves.

How would this fit into the Engine Masters rules?

[ChargerST]

Hi Mike, haven't talked to you in a couple of years. The cfm difference is a possibility. One reason why I was interested in the results is because I saw almost the same results on one of my test engines. In my case I was running a 514 inch Mopar with a 1250 BLP carb and then I switched to Holley EFI. In my case I switched to a port injection setup with a 2000 cfm throttle body. The Holley EFI system had slightly better peak numbers but the torque at the bottom of the range dropped off by 30 or 40 ft-lbs. So very similar results. I never spent much time poking at it since we were looking for other things but I've wondered about it a few times. I assumed it was due to injector placement but I suppose it could be an effect of the much larger throttle body or perhaps I had the injector pulse timed wrong.

https://www.hotrod.com/articles/holley-hp-efi-install/

In this particular case (port injection) do you think it would help to add a 9th injector that sprays into the plenum (close to 100% duty cycle)? This would only add some additional fuel but reduce/remove the dry air in the plenum/runners when port injectors are not on. Basically something like the K&N ECI added to an Edelbrock Pro-Flo 4. I would assume AFRs would be normal and duty cylce for main injectors slightly less than without plenum injector.

[Brian P]

Interesting, if there could be a redo test with 2 different sizes of injectors per cylinder. Then shutdown the larger injector during low RPM.

Yes that might be interesting. The OEM's used that idea for a while but they seemed to have gotten past it so I'm guessing they came up with another solution. They might have developed a more sophisticated control system for the injectors which the aftermarket tuners don't have yet.

Yes. Multipoint port fuel injection. All of the crutches about atomisation and fuel distribution associated with a "wet" manifold go away.

Single-point central fuel injection wasn't around for all that long before multipoint port fuel injection replaced it. I know some manufacturers kept single-point fuel injection around past its sell-by date, but tightening emission regulations eventually forced their hand, and central single-point fuel injection has now been abandoned for a couple of decades.

[hoffman900]

Dual row injectors have been around in racing for just about forever. They work. All your sportbike engines are done this way. These are 200hp/L + applications with full warranties, meet emissions, and can putz around town.

[Mike Laws]

Hi Mike, haven't talked to you in a couple of years. The cfm difference is a possibility. One reason why I was interested in the results is because I saw almost the same results on one of my test engines. In my case I was running a 514 inch Mopar with a 1250 BLP carb and then I switched to Holley EFI. In my case I switched to a port injection setup with a 2000 cfm throttle body. The Holley EFI system had slightly better peak numbers but the torque at the bottom of the range dropped off by 30 or 40 ft-lbs. So very similar results. I never spent much time poking at it since we were looking for other things but I've wondered about it a few times. I assumed it was due to injector placement but I suppose it could be an effect of the much larger throttle body or perhaps I had the injector pulse timed wrong.

https://www.hotrod.com/articles/holley-hp-efi-install/

In this particular case (port injection) do you think it would help to add a 9th injector that sprays into the plenum (close to 100% duty cycle)? This would only add some additional fuel but reduce/remove the dry air in the plenum/runners when port injectors are not on. Basically something like the K&N ECI added to an Edelbrock Pro-Flo 4. I would assume AFRs would be normal and duty cylce for main injectors slightly less than without plenum injector.

Sorry, but I don't know. I've only used the ECI with a carb. It may do what you describe. Except for the cold and hot start features, it only adds fuel based on AFR choice and does nothing for a rich condition.

Thinking more about it, since it uses a dedicated O2 sensor, you would set the AFR to your value and it would inject until reaching that value. So if your port injection produces best power at 13.1 and you set the ECI to 12.5, it will begin wetting the plenum at 12.5, but not saturate because it's only a single or dual injector. You can also set a max RPM point that it would shut the injector(s) off, no matter the AFR. (Basically an upper RPM lean-out.)

It's an interesting thought and I suppose it would work fine in your application.